PROJECT SUMMARY The scientific premise of this proposal is that consumption of highly processed foods contributes to cognitive decline with aging. These dietary effects may lead to health disparities in cognitive decline due to the known greater consumption of highly processed foods in low-income populations. While many factors are associated with the effects of a diet characterized by highly processed foods, including displacement of healthy foods and the myriad of nutrients associated with these, we propose that two nutrients, in particular, may have effects through known or emerging pathways that affect brain function. First, there is evidence that highly processed foods contribute to greater intake of phosphorus (P) through P-based food additives which can increase food P content by ?70%. Vascular risk factors predict cognitive decline and excess P may elevate CVD risk. However, the role of P in cognitive decline remains unclear. One novel pathway is compensatory elevation in fibroblast growth factor 23 (FGF23), a predictor of vascular calcification and CVD. Klotho, a putative anti-aging factor, may be protective against CVD and cognitive decline. Among mice, a high P diet reduced klotho secretion. Excessive P consumption may lower klotho and impair cognitive function through modulation of FGF23, but human data remain sparse. P also elevates parathyroid hormone (PTH), which predicts CVD mortality and presence of cerebral infarcts. Second, vitamin B6 (B6), a critical cofactor for ?100 metabolic reactions, is easily lost with food processing and unlike other B vitamins is not added back to enriched grains. We have shown that low B6 status was associated with inflammation, oxidative stress, metabolic syndrome, diabetes and depressive symptoms in Puerto Rican adults. Earlier work with non-Hispanic white men, showed associations of both diet and plasma B6 (pyridoxl-5'-phosphate,PLP) with 3 y cognitive decline. Latinos appear to be disproportionately burdened by cognitive impairment compared to non-Hispanic whites. Moreover, recent evidence suggests that Puerto Ricans have >twice the odds of low global cognitive function compared to Mexicans. With the growing and aging Latino population in the U.S likely to contribute to major increases in the burden of cognitive decline, it is critical that we improve our understanding of preventive risk factors that may inform interventions. We will use data from the Boston Puerto Rican Health study (BPRHS), a longitudinal cohort study of U.S. mainland Puerto Ricans (45-74 y) with diet, nutritional biomarkers, and cognitive function testing data available at baseline and 2 and 5 y follow-up. We will recruit BPRHS participants to return (~ 8-y from baseline) for a repeat battery of cognitive exams and, for a subset, an MRI scan. We will quantify the associations between processed foods consumption, P status (serum P, FGF23, klotho, and PTH), dietary B6, and PLP, as exposures, with ~8 y cognitive decline (psychometric testing) and brain structure, cerebral infarcts, white matter integrity, and connectivity (MRI) at ~8 y, as outcomes. Further the potential mediative roles of inflammation (CRP, IL-6, and TNF-?) and insulin resistance (HOMA-IR) will be measured.